Ouch! Series resistor please.

You also have a dead short from emitter to emitter of the two transitors once the NPN turns on and has only the Base Emitter of the PNP between it and 30V.

High impedance expected and low impedance base is not working.

I am too tired tonight, but I will try to explain later what I changed from previous circuits and why.

The SCR relay and rest of the circuit I didn't include - just up to the trigger signal.
Use a resistor in series with the SCR trigger because the signal is nearly the full 30V.

 

Thanks to all for your inputs This is really becoming very interesting and the suggestions are very good
I will try some of them this coming week for sure.
What are good common replacements for the 2N3904 and 2N3906 as suggested for higher voltage working.

Just for clarity the signal from the piezo did not switch on the 2N3904 in the diagram I posted much earlier.
Bernard suggested lifting the voltage a bit which seems to make sense to me.

I will try your suggestions this coming week.
Thanks a million

 

Ouch! Series resistor please.

You also have a dead short from emitter to emitter of the two transitors once the NPN turns on and has only the Base Emitter of the PNP between it and 30V.

Good catch Potato Pudding.

Missed that one this morning in a hurry.

 

to Eb:
I miss things all the time.

to Suzuki:
For transistors I would not worry too much.

The only stat that I would worry about is that your transistors have a high voltage rating because of the 30V supply.

Keep currents well under 50mA and running the transistors as switches.

Look for 40V to 50V pnp and npn for under $0.40 each and they should work.

 

Using an emitter follower will give a gate sig. of around .1V. To reliably trigger the SCR , should have 2V or 100μA gate drive. Since timer is floating[?], its ground may be lifted by say .3V by R1 & R2 to make input look like 1.1V peak. Add one 2N3904 as an amplifier, invert to supply a + gate signal.

Bernard,
I tried the circuit you proposed today, but the first 2N3904 is not switching the piezo signal it seems.
Any suggestions, can one use an opamp? Some have suggested using a 555.
Thanks

 

You can't do it with a single 2N3904.

Either you use the circuit that was suggested by Bernard or you try mine, show below. One word of caution though. The 2N3904/06 are rated for 40V only and I would not feel comfortable using it at 30V. I usually uses transistors that rate over 100V in these situation.

I tried Bernards suggestions, but I am still stuck, it seems as if the first transistor does not switch when receiving the piezo pulse. Something ns to be changed and I have no idea. Unfortunately I do not have a scope and the timer runs offf a 1.5V battery.

 

Suzukiman,

The attached is part of a circuit I used to have a 1.5V alarm clock trigger an external circuit (not shown). The buzzer was replaced with R1. I added your circuit after R4.

Ken

 

Suzukiman,

The attached is part of a circuit I used to have a 1.5V alarm clock trigger an external circuit (not shown). The buzzer was replaced with R1. I added your circuit after R4.

Ken

Ken,
Thanks I will try that and let you know if it works
Much appreciated

 

Hi, you can use this diagram to latch any relay

 

Suzukiman,

The attached is part of a circuit I used to have a 1.5V alarm clock trigger an external circuit (not shown). The buzzer was replaced with R1. I added your circuit after R4.

Ken

Ken,
I tried your circuit today, but it seems to work the wrong way around. Q1 is permanently on and keeps the scr switched on. When the piezo pulses are received the scr switches off.
I have the scr switching a led at very low current, so I can see the pulses received by the scr very clearly
Thanks

 

SCR's require a minimum amount of current to remain latch on. What SCR are you using? What sisze resistor are you using in series with the LED?

Ken

 

Ken,
I purposely chose a led and kept the current under 10mA so that the SCR does not latch when the LED comes on. This is just for testing purposes until I have the driver working OK.
When powered up the led will then come on permanently and flash off briefly as the piezo pulses are sent out by the timer. It pulses nicely with the pulses but from on to off.

It seems as if the transistor 2N3906 is switched on permanently on power up and the positive pulses are switching it off on every pulse, which is the exact opposite of what is needed.

 

In the alarm clock I used the buzzer was apparently driven by a NPN open-collector transistor at it's output. One end of the buzzed was always at +1.5V with respect to Common (-1.5V). The other end of the buzzer pulsed between +1.5 and +0.2v. That's why I used the 2n3906 at the output. It could be that you timer is just the opposite. You might need to take the output of the 2n3906 and add a 2n3904 to invert the signal.

Ken

 

In the alarm clock I used the buzzer was apparently driven by a NPN open-collector transistor at it's output. One end of the buzzed was always at +1.5V with respect to Common (-1.5V). The other end of the buzzer pulsed between +1.5 and +0.2v. That's why I used the 2n3906 at the output. It could be that you timer is just the opposite. You might need to take the output of the 2n3906 and add a 2n3904 to invert the signal.

Ken

One side of my timer buzzer is connected to ground and it pulses between ground and +0.8V. Yes I started off with a 2N3904 as per the schematic in my first post and just could not get it to switch. I believed a 2N3904 should switch with the pulses, but obviously I was doing something wrong.
Any comments on my first diagram?

 

May I assume, that referenced to -1.5V battery terminal, one side of the Piezo is +1.5V and the other is 0.0V with no alarm, and pulsing 0.0V/0.8V when in alarm?

Ken

 

May I assume, that referenced to -1.5V battery terminal, one side of the Piezo is +1.5V and the other is 0.0V with no alarm, and pulsing 0.0V/0.8V when in alarm?

Ken

Ken,
One side of the piezo is 0.0V which is connected to the ground of the timer and the negative of the battery(-1.5V). The other side of the piezo has 0V on it until a pulse which then goes from 0V to 0.8V positive.
I hope I have explained it well.
Thanks.

 

Are you observing this on a scope...or a DVM?

Ken

 

Are you observing this on a scope...or a DVM?

Ken

No, I do not have a scope, just using a DVM (multimeter).

 

Sounds like the output driver is an open collector PNP. My guess then is that the non-grounded side of the piezo is actually pulsing between 0.0v and a little less than +1.5v, but your meter is averaging the brief pulses and showing that as 0.8v. So, an NPN should be used instead of a PNP. In your original circuit you were using a 2N3904 in an emitter-follower configuration. The base voltage would have to exceed 1.4v to turn on the SCR. That's 0.6v for the Vbe drop for the transistor plus 0.8v for Vgt drop for the SCR. Better to use a NPN common-emitter followed by a PNP inverter. The attached is just speculation, but may be worth a try.

Ken

 

Ken,
Thanks a million I will try that soon and give feedback.
Much appreciated